TL;DR
- 1.In the only published head-to-head RCT (SURMOUNT-5, NEJM 2025), tirzepatide produced 20.2% vs semaglutide's 13.7% body weight loss at 72 weeks. That is a real, large difference, not marketing.
- 2.Retatrutide, still in Phase 3 trials and not yet FDA-approved, averaged 28.3% weight loss at 80 weeks. 65% of participants on the top dose left the obese BMI category entirely.
- 3.A 2026 Nature study of 27,885 people found GLP1R and GIPR gene variants that predict both how much weight you lose and how much nausea you get, specific to each drug.
- 4.About 10% of people carry PAM gene variants that impair GLP-1 hormone activation itself. These people see roughly half the expected glycemic response on any GLP-1 drug.
- 5.If you are a tirzepatide 'late responder' (under 5% loss at week 12), do not switch yet. A SURMOUNT-1 post-hoc found 90% of late responders still reached meaningful loss by week 72.
In the only direct head-to-head clinical trial, tirzepatide produced 47% more weight loss than semaglutide over 72 weeks. That alone should settle the comparison. Then retatrutide's Phase 3 data arrived with 28.3% average loss at 80 weeks, and a April 2026 study of 27,885 people identified the exact genetic variants that predict who responds to which drug before they take a single dose.
How much more weight tirzepatide-treated patients lost compared to semaglutide in SURMOUNT-5, the first published head-to-head RCT (NEJM, 2025): 20.2% vs 13.7% at 72 weeks. The gap held across all prespecified subgroups. It is large enough to matter clinically for most patients.
All three drugs work through the same family of gut hormone receptors, but each adds another receptor arm. Semaglutide (Ozempic, Wegovy) activates only the GLP-1 receptor to reduce appetite and slow gastric emptying. Tirzepatide (Mounjaro, Zepbound) adds GIP receptor activation, which amplifies fat-cell metabolism and allows faster dose titration because of better GI tolerance. Retatrutide adds a third arm: glucagon receptor activation, which raises resting energy expenditure on top of both appetite and fat-cell effects. The receptor count tracks closely with weight loss outcomes across all three tiers.
Think of it this way. Semaglutide tells your brain "you're full" through one signaling channel. Tirzepatide sends that message through two channels at once. Retatrutide fires all three simultaneously and also tells your body to burn more energy at rest. More channels produce stronger effect, but genetics determines how well each channel transmits your signal.
Tirzepatide vs semaglutide: what the only real head-to-head study showed
For years, the tirzepatide vs semaglutide comparison was fought across separate trials with different patient populations, different baseline BMIs, and different protocols. SURMOUNT-5, published in the New England Journal of Medicine in 2025, was the first direct randomized controlled trial. The results were unambiguous.
Tirzepatide resulted in significantly greater weight reduction than semaglutide, with a mean body-weight loss of 20.2% versus 13.7% at 72 weeks in adults with obesity or overweight. The between-group difference of 6.5 percentage points was consistent across all prespecified subgroups.
SURMOUNT-5, New England Journal of Medicine, 2025
A 6.5-percentage-point difference on a 200-pound person is 13 pounds. On a 250-pound person it is 16 pounds. This is not a marginal pharmacological signal. It is a difference most patients and clinicians would notice over a 72-week protocol. The consistency across subgroups removes the most common objection: that tirzepatide's better outcomes in the original SURMOUNT trials were driven by patient selection or dose optimization artifacts rather than the drug itself.
The mechanism explains the gap. GIP receptor activation, which tirzepatide has and semaglutide does not, amplifies insulin secretion and directly enhances adipocyte (fat cell) lipolysis. The dual receptor engagement also appears to raise the nausea ceiling. Tirzepatide users in SURMOUNT-5 were able to reach and sustain their target doses more reliably, which compounds the efficacy advantage over time.
Should you switch from semaglutide to tirzepatide?
If you have been on semaglutide for at least 16 weeks and response has been limited, the SURMOUNT-5 data supports making the switch. The 6.5-percentage-point efficacy difference is large enough that a population-average non-responder to semaglutide has a meaningfully different expected trajectory on tirzepatide. The caveat: if GIPR genotype is the reason tirzepatide gives some people severe nausea (more on this below), switching requires that awareness.
The late responder rule your doctor might be applying wrong
Standard clinical practice: stop or switch if you have not hit 5% weight loss by week 12. A post-hoc analysis of SURMOUNT-1, published in Diabetes, Obesity and Metabolism in 2025, found something that directly challenges this protocol. Eighteen percent of tirzepatide patients were "late responders," meaning under 5% loss at week 12. Ninety percent of those late responders still crossed the 5% threshold by week 72, with a median crossing around week 25 and a mean final loss of 11%. The "stop at 12 weeks" rule may be pulling people off a drug that would have worked for them, just more slowly. This finding is specific to tirzepatide. Whether the same late-responder pattern exists for semaglutide has not yet been established in published data.
Where does retatrutide fit? The 28% data and what it means
Retatrutide is not FDA-approved and is not available through compounding. That disclaimer matters. But the TRIUMPH-1 Phase 3 results, published in the New England Journal of Medicine in May 2026, are the most significant weight-loss pharmacology data in a decade.
Average weight loss in TRIUMPH-1 Phase 3 at 80 weeks on retatrutide 12 mg (NEJM, May 2026). At 104 weeks in the extension, the figure reached 30.3%, placing it within the range of bariatric surgery outcomes. 65.3% of participants on the top dose left the obese BMI category by the end of the trial.
The TRIUMPH-1 data: 2,339 patients, 80 weeks on the 12 mg dose. Mean weight loss 28.3%. At 104 weeks: 30.3%. Individually, 45.3% of participants on the top dose achieved 30% or more weight loss. And 65.3% dropped out of the obese BMI category entirely. For context, published bariatric surgery outcomes typically range from 25-35% excess weight loss. Retatrutide is touching those numbers with a weekly injection.
A body composition substudy from Lancet Diabetes and Endocrinology in 2025 addressed the muscle-loss concern: roughly 75-80% of weight lost on retatrutide was fat mass, not lean mass. This ratio is similar to semaglutide and tirzepatide outcomes, despite the dramatically higher total weight loss. The fear that triple receptor activation causes disproportionate muscle wasting does not appear in the current published data.
Regulatory timeline: Eli Lilly is expected to file an NDA in late 2026. FDA approval is not anticipated before Q1-Q2 2027. There is no compounding pathway currently available. Anyone representing access to retatrutide right now is not selling what they claim.
How your GLP1R and GIPR genes predict which drug works for you
In April 2026, the 23andMe Research Institute published the largest GLP-1 pharmacogenomics study ever conducted in Nature, analyzing 27,885 people who had taken semaglutide or tirzepatide. Two findings directly inform the three-way comparison here. If you already have a 23andMe or AncestryDNA file, both variants are potentially covered in your raw data. See the semaglutide non-responder genetics article for the full variant breakdown.
The T allele at this position was associated with approximately 0.76 kg of additional weight loss per allele, reaching genome-wide significance (p = 2.9 x 10^-10). Carriers of two copies lost about 3.3 pounds more over the median treatment window than non-carriers on identical regimens. This effect applies across semaglutide and tirzepatide. Favorable GLP1R genotype predicts stronger response to GLP-1 receptor agonism in general, not just one drug.
The same 2026 Nature GWAS identified GIPR variants that specifically predict nausea and vomiting on tirzepatide, and not on semaglutide. Because tirzepatide activates GIP receptors and semaglutide does not, the pharmacological logic is direct: the nausea signal is coming from the GIP receptor arm. If you tolerated Ozempic well but Mounjaro was rough, your GIPR genotype may explain exactly why. This is the first genetic-level explanation for a pattern many patients had already noticed, without a mechanism to point to.
A Stanford and ETH Zurich study published in Genome Medicine in April 2026 identified PAM gene variants (p.S539W and p.D563G) carried by roughly 10% of the population that impair GLP-1 amidation, the enzymatic step that activates the hormone before it can bind any receptor. Non-carriers hit HbA1c targets in about 25% of cases at 6 months on treatment. p.S539W carriers hit them in 11.5% of cases. This applies to all three drugs because the bottleneck is upstream of receptor activation. No dose increase on any GLP agonist fully overcomes impaired hormone amidation.
The combined picture from the 2026 genetics research: GLP agonist response exists on two layers. A polygenic layer (GLP1R and GIPR variants, modest individual effects, acting at the receptor) and what looks like a discrete monogenic layer (PAM variants, large effect in roughly 10% of people, acting upstream at hormone synthesis). Neither layer is currently tested before prescription in standard practice. Both are potentially visible in consumer genetic data you may already have.
How retatrutide changes the genetics calculus
GLP1R genotype matters most when a drug works primarily through the GLP-1 receptor. Retatrutide adds GIPR and GCGR activation as independent pathways. For someone with unfavorable GLP1R genotype, this is pharmacologically meaningful: if the GLP-1 receptor arm is inherently less responsive, the glucagon and GIP arms can contribute weight loss through pathways not constrained by GLP1R variant effects.
No published head-to-head trial between retatrutide and tirzepatide exists yet. Weill Cornell is running one, with results expected in December 2026. But the mechanism argument is coherent: if you are a weak GLP1R responder and tirzepatide worked better than semaglutide for you, retatrutide may add another tier still, with the glucagon arm compensating for whatever GLP-1 receptor weakness remains. The full pharmacogenomics picture for retatrutide will not be clear until GCGR variant data is analyzed from the TRIUMPH trial population. For now, the strongest available option with robust efficacy data in approved drugs is tirzepatide. Read the full retatrutide deep-dive for the trial details.
| Drug | Receptors | Avg Weight Loss | Source | FDA Status (2026) | Key Gene Predictors |
|---|---|---|---|---|---|
| Semaglutide (Ozempic / Wegovy) | GLP-1R only | 13.7% | SURMOUNT-5, NEJM 2025 | Approved (T2D 2021, obesity 2021) | GLP1R, PAM |
| Tirzepatide (Mounjaro / Zepbound) | GLP-1R + GIPR | 20.2% | SURMOUNT-5, NEJM 2025 | Approved (T2D 2022, obesity 2023) | GLP1R, GIPR, PAM |
| Retatrutide | GLP-1R + GIPR + GCGR | 28.3% | TRIUMPH-1, NEJM May 2026 | Phase 3, not approved (NDA 2026) | GLP1R, PAM (GCGR data pending) |
Which GLP agonist actually matches your situation?
The genetics research is not yet routinely applied before prescription. But if you have a 23andMe, AncestryDNA, or similar file, the relevant variants may already be in your data. The GLP1R gene prediction analysis explains the full variant table, and the DNA-first decision framework shows how GLP1R and GIPR fit into a broader metabolic protocol.
If you have not tried any GLP agonist: the SURMOUNT-5 data makes a clear case for tirzepatide over semaglutide if you have access and no contraindications. The 6.5-point difference held across all subgroups. Semaglutide remains appropriate when cost or access is the controlling factor.
If semaglutide produced limited results: two genetic explanations now have names. Unfavorable GLP1R genotype reduces receptor-level response. PAM gene variants reduce the activated hormone upstream of all receptor effects. Tirzepatide adds a second receptor mechanism that partially bypasses the GLP1R limitation, which is why some people who plateau on semaglutide do meaningfully better on tirzepatide.
If tirzepatide gave you severe nausea: the 2026 Nature GWAS identified GIPR variants as the most likely pharmacogenomic cause. Semaglutide does not activate GIP receptors, so switching back removes that specific side effect trigger. Alternatively, a dose reduction and slower titration on tirzepatide is supported by the late-responder data: lower steady-state doses still produce significant weight loss in GIPR-sensitive patients who cannot tolerate the top dose.
If you carry PAM gene variants: all three drugs are working with impaired hormone activation upstream of all receptors. This population sees smaller absolute effects from every GLP agonist, but the drugs still work at meaningful levels. Dose optimization (reaching and sustaining the target dose without early discontinuation) matters more in this population, not less, because the upstream bottleneck means greater downstream stimulation is needed to compensate. Lifestyle interventions that operate independently of GLP-1 pathways carry more relative weight here.
Compounding access update: semaglutide compounding under 503B outsourcing facilities ended in mid-2025 following the FDA shortage resolution. The FDA proposed in April 2026 to permanently remove semaglutide and tirzepatide from the 503B bulk substances list. Any compounded product is now legally precarious. Use only FDA-approved branded products or patient-specific 503A compounding through a licensed pharmacy with a valid prescription.
Verdict: Tirzepatide is the evidence-based first choice over semaglutide right now, with a 6.5-point head-to-head advantage across 72 weeks in a direct RCT. Retatrutide sets a new ceiling at 28% but is not yet available. Your GLP1R and GIPR genotype predicts how strongly you respond to each agonist tier, and about 10% of people carry PAM variants that reduce GLP-1 hormone activation upstream of all three drugs. Upload your genetic data to see which agonist tier your biology actually responds to, or order a DNA kit to get a personalized GLP-agonist match before spending months on the wrong drug.
Your DNA shapes how you respond to each of these.
A personalized report scores 25+ peptides against your unique genetic profile — including the ones covered in this article.
Frequently asked questions
Is tirzepatide better than semaglutide for weight loss?
Yes, based on the only direct head-to-head randomized controlled trial published to date. SURMOUNT-5 (NEJM, 2025) found tirzepatide produced 20.2% vs 13.7% weight loss at 72 weeks. The 6.5-percentage-point gap was consistent across all patient subgroups. How much better tirzepatide works for you individually depends partly on your GLP1R and GIPR genotype.
What is retatrutide and when will it be FDA-approved?
Retatrutide is a triple GIP, GLP-1, and glucagon receptor agonist currently in Phase 3 trials. TRIUMPH-1 (NEJM, May 2026) showed 28.3% average weight loss at 80 weeks, the highest ever recorded in a pharmaceutical weight-loss trial. Eli Lilly is expected to file an NDA in late 2026. FDA approval is not anticipated before Q1-Q2 2027. It is not currently available through compounding.
Can 23andMe data tell me which GLP-1 drug will work best for me?
Potentially yes. A 2026 Nature study of 27,885 people identified GLP1R and GIPR variants that predict both weight loss magnitude and nausea risk on tirzepatide vs semaglutide. Your 23andMe or AncestryDNA raw data file may contain these variants. No commercial service currently runs this analysis as a standard product, though 23andMe's Total Health includes a GLP-1 response report for subscribers.
Why did I lose less weight on semaglutide than my friend on the same dose?
The most pharmacologically grounded explanations are GLP1R genotype and PAM gene status. GLP1R rs10305420 carriers lose meaningfully more weight per allele. PAM gene variants, carried by roughly 10% of people, impair the enzymatic step that activates GLP-1 before it reaches any receptor. Both of these biological differences produce large, real variability on the same drug at the same dose.
What is the PAM gene and how does it affect GLP-1 drugs?
PAM (peptidylglycine alpha-amidating monooxygenase) is an enzyme that activates GLP-1 through a chemical process called amidation. Variants in PAM, carried by about 10% of people, reduce this activation step. A Stanford study in Genome Medicine (April 2026) found that PAM variant carriers hit their blood sugar targets in 11.5% of cases on GLP-1 drugs, compared to 25% for non-carriers. This bottleneck applies to all three GLP agonists because it is upstream of receptor binding.
Should I stop tirzepatide if I have not lost weight after 12 weeks?
The standard 12-week cutoff may be wrong for tirzepatide specifically. A post-hoc analysis of the SURMOUNT-1 trial (Diabetes, Obesity and Metabolism, 2025) found that 90% of tirzepatide patients who were under 5% weight loss at week 12 still reached meaningful weight loss by week 72, with a median response appearing around week 25. Discuss this data with your clinician before discontinuing, especially if you have not yet reached the target maintenance dose.
Why did tirzepatide make me so nauseous when Ozempic did not?
The most likely explanation is your GIPR genotype. The 2026 Nature GWAS of 27,885 people found that GIPR variants specifically predict nausea on tirzepatide and show no equivalent effect for semaglutide. Because tirzepatide activates GIP receptors and semaglutide does not, the GIP receptor arm is the source of the differential nausea signal. Slower dose titration on tirzepatide reduces but does not eliminate this for people with the relevant GIPR variants.
This article is for informational and educational purposes only. It is not medical advice and does not diagnose, treat, cure, or prevent any disease. Consult a qualified healthcare professional before starting any peptide protocol. Individual results vary.